-
Notifications
You must be signed in to change notification settings - Fork 1
Compound Mind Patterns
Full reference for L3 Compound Mind — 8 solution axes and 8 cross-domain patterns. Load when cross-domain synthesis is needed, or when building a domain skill that benefits from patterns in other fields.
| Axis | Question | When most powerful |
|---|---|---|
| Additive | What can be added? | When resources are available and the gap is a capability gap |
| Subtractive | What can be removed? | Often faster than adding. Removes dead weight identified in L2. |
| Multiplicative | What can be scaled? | When a working small solution exists and the constraint is reach |
| Division | What can be split or outsourced? | When complexity is the bottleneck, not capability |
| Substitutive | What can be swapped for something better? | When the current input is the constraint |
| Reframing | Is the target itself wrong? | When all solutions keep failing — the problem may be wrong |
| Temporal | Can timing change the outcome? | When the problem is cyclical or opportunity-dependent |
| Relational | Can a partnership or relationship solve this? | When the required capability exists elsewhere |
Map all axes internally. Test before including in output:
"Would knowing this alternative path change the decision? If no → omit."
The goal is to understand all paths and transmit only what matters. Dense output over complete output.
- L3-Compound-Mind — Pipeline summary
- L2-System-Lens — Leverage point from L2 focuses L3 axis priority
- Ariya4 — Path field in Ariya4 is where L3 optimal path feeds
# Compound Mind Patterns
[← Back to Home](Home)
Full reference for L3 Compound Mind — 8 solution axes and 8 cross-domain patterns. Load when cross-domain synthesis is needed, or when building a domain skill that benefits from patterns in other fields.
| Axis | Question | When most powerful |
|---|---|---|
| Additive | What can be added? | When resources are available and the gap is a capability gap |
| Subtractive | What can be removed? | Often faster than adding. Removes dead weight identified in L2. |
| Multiplicative | What can be scaled? | When a working small solution exists and the constraint is reach |
| Division | What can be split or outsourced? | When complexity is the bottleneck, not capability |
| Substitutive | What can be swapped for something better? | When the current input is the constraint |
| Reframing | Is the target itself wrong? | When all solutions keep failing — the problem may be wrong |
| Temporal | Can timing change the outcome? | When the problem is cyclical or opportunity-dependent |
| Relational | Can a partnership or relationship solve this? | When the required capability exists elsewhere |
Subtractive is systematically underused. Most problem-solving defaults to additive solutions (add a feature, add a process, add a person). Subtractive solutions — removing the constraint, removing the complexity, stopping the activity — are often faster, cheaper, and more durable. L3 must always explicitly consider the Subtractive axis.
These patterns appear across multiple domains. When a problem matches a pattern, the pattern's insight applies directly.
Origin: Control systems engineering Appears in: Markets, biology, software architecture, coffee roasting, organizations
Structure: Output → Sensor → Comparator → Actuator → Output (loop)
Insight: The sensor + comparator combination is the leverage point. Improving the output directly (the Additive axis) is less effective than improving what measures it and compares it to target (the Substitutive or Subtractive axis applied to the feedback mechanism).
Application: When a system keeps underperforming despite intervention, check whether the feedback loop itself is measuring the right thing. A system optimizing for the wrong metric will compound in the wrong direction.
Origin: Diffusion of innovations (Rogers) Appears in: Technology adoption, startups, skill acquisition, epidemics, market penetration
Structure: Slow start → exponential growth → saturation plateau
Insight: The optimal strategy changes at each phase. Early phase: find the early adopters (not the mass market). Growth phase: remove friction (not add features). Saturation phase: segment (not scale). Position on the curve determines the correct axis, not the goal.
Application: Many strategies fail because they apply growth-phase tactics to early-phase problems, or early-phase tactics to saturation-phase markets.
Origin: Manufacturing optimization (Goldratt) Appears in: Software development, supply chains, learning systems, team performance
Structure: System output = throughput of the weakest link (constraint)
Non-bottleneck improvement = zero gain in system output
Insight: Find the single binding constraint. Apply all improvement effort there. After it is resolved, a new constraint emerges — find it and repeat. Improving non-constraints is waste.
Application: In L2 System Lens, the leverage point IS the constraint in constraint theory terms. L3 Compound Mind should apply improvement axes to the leverage point first — not to the most visible or most comfortable problem.
Origin: Economics (Akerlof — "Market for Lemons") Appears in: Markets, hiring, negotiations, medical diagnosis, investor relations
Structure: One party has more information than the other.
The less-informed party misprices risk.
The more-informed party exploits the misprice (or: the market fails).
Insight: Map who holds what information. The party with less information is exposed — either to exploitation or to poor decisions. The leverage point is the information gap itself, not the behavior it produces.
Application: In L4 Interest Map — the structural incentive that produces biased data is often an information asymmetry. The party generating the data knows more than the party consuming it.
Origin: Thermodynamics Appears in: Market regime changes, cultural shifts, political revolutions, technology disruptions
Structure: Linear behavior below threshold → discontinuous jump at threshold → new linear behavior
(ice → water → steam: the transition is not gradual within a phase)
Insight: Linear extrapolation fails near a phase transition threshold. Historical data from within-phase behavior does not predict through-threshold behavior. When a system is near a threshold, the relevant question is: which side of the threshold will we be on? — not how fast is the current trend continuing?
Application: In market analysis, regime changes are phase transitions. The SMC Layer's Decision Zone (L3) maps to this pattern — it is the price level near a structural threshold.
Origin: Finance (compound interest) Appears in: Skill development, relationships, brand equity, code quality, reputation, distribution
Structure: Return N applied to a larger base than Return N-1
Result: exponential growth from linear input
Insight: Small early advantages compound over time. The asymmetry: a 1% daily improvement compounds to 37× over a year; a 1% daily decline compounds to 0.03×. Time is the most systematically undervalued variable in strategic decisions.
Application: In Path selection (Ariya4 Field 4), always prefer the path that creates a compounding return over the path with a larger immediate return but no compounding mechanism.
Origin: Pareto / economics Appears in: Wealth distribution, code defects, customer revenue, social reach, market returns
Structure: ~20% of inputs → ~80% of outputs (approximately; ratios vary)
Insight: Finding and serving the 20% produces more output than optimizing the 80%. This applies to: customers (top 20% generate ~80% of revenue), defects (20% of code = 80% of bugs), effort (20% of decisions = 80% of outcomes).
Application: In L3 Compound Mind, when evaluating Multiplicative and Additive axes — apply to the 20%, not the 80%.
Origin: Statistics (Galton) Appears in: Athletic performance, investment returns, exam scores, sales performance, medical outcomes
Structure: Extreme observations → less extreme observations — independent of intervention
Insight: Many "successful interventions" are regression to mean in disguise. The high-performer who received coaching improved — but would have regressed regardless. The poor-performer who was replaced improved — but the replacement may also regress. Do not attribute regression to intervention.
Application: In L4 Blind Spot — when evaluating whether an intervention worked, check whether the sample was selected for extreme performance. If yes, regression is a competing explanation for improvement.
When analysis is stuck cycling on the same axis:
| Protocol | Instruction |
|---|---|
| Axis Rotation | "I am on [X] axis. What does the [Y] axis solution look like?" |
| Domain Transplant | "How does [different field] solve an analogous problem?" |
| First Principle Reset | "What is the actual goal? Am I solving the right problem, or a convenient one?" |
| Inversion Path | "What would make this worse? Now reverse that — is that the solution?" |
Map all axes internally. Test before including in output:
"Would knowing this alternative path change the decision? If no → omit."
The goal is to understand all paths and transmit only what matters. Dense output over complete output.
- L3-Compound-Mind — Pipeline summary
- L2-System-Lens — Leverage point from L2 focuses L3 axis priority
- Ariya4 — Path field in Ariya4 is where L3 optimal path feeds
[← Back to Home](Home)